Calculate Portfolio Variance in Panel Data

[Nabil Alkafri]

Hello everyone,

I have a dataset which looks as follows:

Date crsp_fundno cusip stock_return stock_weight

1993-01 1 1111 0.05 0.40
1993-01 1 1112 -0.02 0.20
1993-01 1 1113 0.07 0.40
1993-02 1 1111 0.03 0.33
1993-02 1 1112 0.04 0.33
1993-02 1 1113 -0.02 0.33
1993-01 2 1111 0.05 0.10
1993-01 2 1112 -0.02 0.20
1993-01 2 1113 0.07 0.40
1993-01 2 1114 0.03 0.15
1993-01 2 1115 -0.01 0.05
1993-01 2 1116 0.01 0.10
1993-02 2 1111 0.01 0.15
1993-02 2 1112 -0.05 0.12
1993-02 2 1113 0.04 0.30
1993-02 2 1114 0.02 0.20
1993-02 2 1115 -0.05 0.10
1993-02 2 1116 0.03 0.10
.....

I would like to calculate the historical portfolio variance for every portfolio at every month.

Can someone please elaborate as to how can I calculate that?

Thank you very much in advance.

Best,
Nabil

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

If you have an identifier for portfolios, you can do it with egen monthlysd=sd(stockreturn) , by(date fundno)

If you want weights, you may need to weight the returns before doing the egen. However, I'm not sure exactly what the standard deviation for a portfolio in a given month means.